PSI - Issue 27

Aditya Rio Prabowo et al. / Procedia Structural Integrity 27 (2020) 171–178 Prabowo et al. / Structural Integrity Procedia 00 (2019) 000 – 000

172

2

1. Introduction Since a long time ago, the collision phenomenon has become remarkable with accidents both on the sea and land. Limitless possibility of the scenario and cause, make the investigation and research on this field was done continuously until the 20th century. The tragic accident of Royal Merchant Ship Titanic on Atlantic Ocean 1912 is a famous example of this phenomenon. This accident made the related parties perform investigation and research to make rules and safety codes on the sea to avoid the same tragedy in the future. The phenomenon process in ship collision is divided into two main parts, i.e., external dynamics and internal mechanics of ship collision, which both of them interrelate with each other. The method to assess this phenomenon itself has experienced development and improvement up to this day. In order to get a satisfactory result in collision analysis, Kitamura (2002) from Japan stated in his previous work that the investigation should be based on real accident, experiment, or finite element method (FEM) approach. Previous works on ship-ship collision were performed by Haris and Amdahl (2013) and Prabowo et al. (2017); the interaction between ice and structure by Gao et al. (2015) and Bae et al. (2016); and statistical data on ship collision by Pedersen and Zhang (2000) and Khan et al. (2020). These works were performed by the FEM approach and real collision data. The FEM implementation and deployment are preferred to these days as the growth progress of computation instruments. Numerical data can present reliable results if the model is defined properly. In this case, the accident or experiment is needed to verify the model itself (Calle et al., 2017; Prabowo et al., 2018; Ma et al., 2020). In this present work, the authors present the study on the ship collision phenomenon with collaboration between the finite element approach and real collision accident data. The research will be focused on the energy dissipation of collisions in several impact scenarios. The scenarios will be made based on different locations of the target point on the side structures and material model based on selected ship material types. 2. Virtual simulation In the last two decades, the human race has undergone significant development, especially in computational technology. This technology was applied in various aspects of human life, such as transportation, production as well as calculation and measurement field. In the engineering field itself, the elements of calculation and measurement itself hold a vital role as the prediction instrument of behavior and characteristic of observed objects, which can be performed with virtual simulation. The simulation can be considered as an alternative way to solve engineering problems since the failure probability and cost, in some cases, cannot be controlled and immense. The application of the virtual simulation was also deployed in naval architect, marine, and ocean engineering to assess some physical phenomenon. In collision phenomena, the finite element method (FEM) approach can be considered as a popular method. In FEM analysis, two methodologies exist to solve engineering problems based on their types. This methodology, namely implicit and explicit techniques. For computers, matrix multiplication is not difficult. Matrix inversion is the more computationally expensive operation. In the implicit methodology, the solution is obtained by inverting the stiffness matrix, and this operation is the CPU-intensive operation. The displacement vector will be solved in the following stages. The explicit methodology produces the result by solving the acceleration vector in the first stage. In this methodology, the mass matrix will be considered as lumped. This process makes the computational process less intensive than implicit. As per time integration, implicit is unconditionally stable, whereas explicit time integration is not because, in explicit methodology, the constraint in terms of critical steps should be fulfilled. This situation makes explicit techniques require much tinier time steps and very small steps. Assessment and analysis of complex phenomena and involve high nonlinearity as well as require tiny steps, such as collision and grounding, the explicit technique is most suitable. The required calculation efforts are fewer than the commonly used implicit methods. Converge of calculations is much easier to realize. Computational code based on explicit techniques includes ABAQUS, DYTRAN, ANSYS, and LS-DYNA. Previous researchers had performed an investigation on the physical phenomenon by using FEM analysis before. The conclusion of their researches is that element type and mesh fineness becomes two important factors in consideration of structural modeling by FEM. Investigations indicated that a very large number of elements are required to obtain accurate results for components that experienced direct contact with each other.